Potential economic impact of using a proprietary willow bark extract in outpatient treatment of low back pain: an open non-randomized study.

An open, non-randomised, study (postmarketing surveillance) was carried out on three groups of patients aged 18 to 80 presenting over an 18 month period with acute exacerbations of low back pain. The objective was to assess the possible economic impact of including a regular dose of proprietary willow bark extract (Assalix) in the treatment provided. A first group of 115 patients, presenting to 3 general practitioners in the first 3 months, was prescribed a daily dose of extract containing 120 mg of salicin (group W120). They also had access, if necessary, to the range of conventional treatments allowed for in the general practitioners' budgets. A second group of 112 patients presenting to the same general practitioners over the next 15 months, was prescribed extract equivalent to 240 mg salicin per day (group W240). A third "control" or "comparator" group of 224 patients, presenting to 3 orthopedists (specialists in physical medicine) over the whole 18 month period, received only the conventional therapeutic options allowed in the orthopedists' budgets (Group C). In the group C patients, the exacerbations had been shorter but the pain had been more intense as judged by Arhus Index and Total Pain Index. After 4 weeks of treatment, about 40% of group W240 patients were free of pain whether or not they had to resort to supplementary treatments. In group W120 as a whole, about 19% of patients were pain-free at 4 weeks, but only 8% of those who did not resort to supplementary treatment. In group C, 18% of patients were painfree. These findings were reflected reasonably well in the changes in the Arhus Index and Total Pain Index, and the findings in group W240 were consistent with those in a previous randomised controlled trial. Multivariable modelling to examine for possible confounding effects tended to identify membership of group W240 as an independent explanator of better pain relief than membership of group C. Though the measures of effect tended to be similar in group W120 as a whole and group C, the avoidance of more expensive conventional treatments in group W120 meant that the average cost per patient of treatment was reduced by about 35-50% (health service and private costings respectively). The better pain relief in group W240 was accompanied by an even smaller reliance on supplementary conventional treatments than in group W120 but the extra savings on these were outweighed by the extra cost of the additional Assalix so that the average cost per patient was reduced by 14-40% of the costs in group C. The possibility is discussed that, if orthopedists had relied more on regular full dosing with NSAIDs, they might have increased the effectiveness and reduced the cost of their treatment, though with the possibility of more side effects. Substituting established NSAIDs with COX-2 inhibitors might reduce the side effects, but at greater cost than with the Assalix.